An extract, transform, and load (ETL) system refers to a system that processes data by extracting the data from an external source (e.g., database), transforming the data to fit the operational needs of a target system, and loading the data onto the target system (e.g., a data warehouse). An ETL system is scalable if the ETL system is able to handle varying volumes of data by proportionally allocating lesser or additional resources to handle the data either at the source or the target.
In most ETL systems, concurrent extraction of data from a source and further management of the data for concurrent transformation and loading is not supported. The above results in a bottleneck effect which leads to inefficient management of the data across the entire system.